Göran Selander [email protected]<mailto:[email protected]>
wrote:
>We could e.g. define a COSE Header Parameter (say 'hpke-alg'; label 11)
>signifying that this is an HPKE algorithm, in which case e.g. {11 : 17} would
>indicate DHKEM(P-384, HKDF-SHA384).
A new 'hpke-alg' parameter with values from the “HPKE KEM Identifiers” IANA
registry seems like a very good solution. I think this makes a lot of sense in
this case. All future HPKE KEMs would then automatically be usable in COSE and
we would avoid COSE WG work registering new HPKE KEMs.
If COSE adopt this draft, we should discuss if COSE WG sees HPKE as the main
way to use KEMs in COSE in the future. Note that all the current ECDH-ES
algorithms are KEMs. A general HPKE mechanism would mean COSE can automatically
use any HPKE PQC KEMs, but nothing would stop COSE from registering more
optimized PQC KEMs if needed.
Cheers,
John
_______________________________________________
COSE mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/cose
